Review of Guillaume, Antoine: A Grammar of Cavineña. Katharina Haude To cite this version: Katharina Haude. Review of Guillaume, Antoine: A Grammar of Cavineña.. Studies in Language, John Benjamins Publishing, 2011, pp.913-921. halshs-00692111 HAL Id: halshs-00692111 https://halshs.archives-ouvertes.fr/halshs-00692111 Submitted on 27 Apr 2012 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Guillaume, Antoine. 2008. A Grammar of Cavineña. Berlin/New York : Mouton de Gruyter [Mouton Grammar Library 44]. xxxiv+900pp. (ISBN 978-3-11-018842-4) Reviewed by Katharina Haude (SeDyL/CELIA, CNRS) Cavineña is an endangered language of northern Bolivia with approximately 1,200 speakers. Together with the neighbouring languages Ese-Ejja, Araona, Reyesano, and Tacana, it forms the Tacanan family, which has sometimes been classed with Panoan. Antoine Guillaume (henceforth AG) has carried out research on Cavineña since 1996, and A Grammar of Cavineña, a revised version of his 2004 doctoral dissertation, is the first comprehensive description of the language. It is almost exclusively based on natural discourse data collected by the author during a total of about 15 months of fieldwork, complemented with material from the SIL linguists Elizabeth L. Camp and Millicent R. Liccardi, who studied the language from the early 1960s to the mid-1980s. The grammar is written in a functional-typological style. The volume contains 20 chapters of grammatical description, two glossed texts (ca. 20 pages), a bidirectional Cavineña-English word list (ca. 40 pages), a list of affixes, and a subject index. It is organized according to the syntactic components of the language, zooming in from overviews of larger units to detailed discussions of their segments: a general grammatical overview (Chapter 4) is followed by two large sections on the predicate (Chapters 5-11) and on the noun phrase (Chapters 12-15), each introduced in turn by overview chapters. Additional chapters deal with phonology, postpositions, particles, and subordinate clauses. Chapter 1 (“The language and its speakers”) provides geographical, sociolinguistic, cultural, and historical background information. The Cavineñas live in small communities along the right bank of the Beni river, between the towns of Reyes in the South and Riberalta in the North, a region that seems to coincide largely with their ancestral settling area. Hardly anything is known about the Cavineñas’ original culture and beliefs. In the late 18th century, they were contacted by Spanish Franciscans, who converted them to Catholicism and stayed until 1941; after this, American Catholics replaced them. In the 1950s, when the SIL started working in the region, part of the population converted to Protestantism. The impact of the rubber boom was relatively limited, with “only” some Cavineña families sold by the priests to a rubber company in 1940. 1 The chapter gives a nice charactization of the Cavineña sense of humour and their habit of nickname-giving. It also provides information on previous work on the language, fieldwork and data collection, the nature of the data corpus, the writing system, as well as on typological characteristics of Cavineña. It is clear right from the beginning that the grammar is based on very intensive fieldwork during which AG got to know the language and its speakers very well. The way in which sources and contexts of the examples are transparently provided throughout the book, fully confirm this impression. Chapter 2 (“Phonology”) shows that with its small phoneme inventory (4 vowels, 20 consonants, neither /f/ nor /g/) and a simple syllable structure (no closed native syllables), the phonological properties of Cavineña are typical of the South-West Amazon area. The glottal stop is not a phoneme, but can occur to resolve a hiatus. Pitch accent marks word boundaries. Chapter 3 lays out the distinction between grammatical and phonological words, based on Dixon & Aikhenvald (2002). A Cavineña phonological word must have more than one syllable. Grammatical words can be either free or cliticized. Clitics or clitic combinations can have their own stress domain and can thereby form phonological words of their own. Chapter 4 (“Grammatical overview”) provides basic information on the language. Most topics are treated in depth in the subsequent chapters, but some (main-clause structure, coordination) are not, so this is not a chapter to be skipped. Cavineña has free constituent order. Grammatical functions are encoded by case marking, the core cases being ergative (suffix/enclitic -/=ra) for A and absolutive (generally unmarked) for S/O. Syntax is nominative-accusative. Chapters 5 through 11 deal with predicate structure, Chapter 5 providing an overview (partly repeated from Chapter 4). It presents the predicate as consisting of 11 morpheme slots (A-K). Some slots interact with each other, e.g. the first and the last, of which either one or both have to be filled by an inflectional affix. Some can contain more than one morpheme, e.g. the slot that hosts the “Aktionsart” suffixes. A predicate can only be headed by a verb; copula clauses are considered a different clause type, with inflectional morphology carried by a copula verb (ju- ‘be’). Two classes of verbs can be distinguished: inflecting and non-inflecting. Inflecting verbs are inherently either transitive or intransitive, as is reflected by their morphological and syntactic properties. Four verbs (‘give’, ‘tell’, ‘ask’, ‘take away’) are ditransitive and take two absolutive arguments. Inflecting verbs can be derived from other word classes, but underived inflecting verbs form a closed class. 2 Non-inflecting verbs require an auxiliary, either a- ‘affect’ (transitive) or ju- ‘be’ (intransitive, identical with the copula), to carry any inflectional morphology. This class is open and freely incorporates borrowings from Spanish, especially nouns (e.g. viaje ju- ‘travel’, from Spanish viaje ‘journey’). It also contains denominal verbs derived through conversion, the conversion being obvious from the shift in meaning: for instance, the intransitive verb ‘fish’+ ‘be’ does not mean ‘to be a fish’, but ‘to fish (with an arrow)’, with the agent as the S argument; the transitive verb ‘bait’ + ‘affect’ does not mean ‘to make the bait’ but ‘to put O as bait’. The non-inflecting verb class is extremely interesting and calls for further research from a typological perspective, e.g. in the context of light-verb constructions in languages like English or in comparison to other languages with a closed class of verbs (e.g. Jaminjung; Schultze-Berndt 2000). Chapter 6 presents the inflectional affixes. They mark tense/aspect/modality and share two affix slots (A and K), in which they are mutually incompatible. In particular contexts (predictability, repetition), often accompanied by marked intonation, inflectional affixes can be absent (but see Chapters 10 and 12, respectively, for the omission of inflectional morphology deriving non-inflecting verbs and nouns). Chapter 7 describes “Aktionsart suffixes”. The term Aktionsart, generally used to refer to the temporal/aspectual properties of a verb, is taken here by its literal meaning ‘kind of action’. The suffixes in this group cover a broad range of meanings, including aspects (completive, incompletive, durative), but also motion (distinguishing between stable and unstable targets), time of day, emotions, or participant number. While assigned to a single morphological slot, two or three Aktionsart suffixes can co-occur. Chapter 8 deals with valency-changing mechanisms, which are carried out by verbal affixes, root reduplication, and auxiliary change. The valency-reducing mechanisms, which only apply to transitive (and occasionally ditransitive) verbs, are passive, reflexive/reciprocal, and antipassive. The two instances of the passive suffix, -ta and -tana, are tentatively analysed as free variants of one morpheme (a revised analysis can be found in Guillaume to appear) that marks agentless passives and anticausatives. The passive is convincingly argued to originate from a third-person plural morpheme. The reflexive/reciprocal, marked by a circumfix, derives intransitive verbs whose single argument represents either agent and patient simultaneously or just the agent, with a focus on the activity itself. The antipassive is derived from inflecting verbs through reduplication. The derived verb is intransitive verb with only an A argument, the O argument being understood but not 3 expressed. The verb denotes a “culturally identified activity” (p. 278) like ‘eat (intr.)’, ‘row’ (from ‘stir’), ‘have drinks’. Similar to the reflexive/reciprocal, the focus is on the activity itself. The antipassive of non-inflecting verbs is formed by “exchanging the transitive auxiliary a- ‘affect’ for the intransitive auxiliary ju ‘be’” (p. 282f.), illustrated with the verbs ‘say yes to’ > ‘agree’; ‘tell (sth.) to’ > ‘talk (with sb.)’; ‘continue to’ > ‘continue’ (p. 283f.). It is not entirely clear to me why this is called “derivation”, since the subsititution